KUTSCHALE: LOW-FREQUENCY PROPAGATION IN THE ICE-COVERED ARCTIC OCEAN 



corresponds to incoherent waves reflected from the ocean floor. An 

 important point here is that the effects of ice roughness on propaga- 

 tion loss appear to be nearly negligible for waves below 10 Hz so that 

 these waves will travel to very great ranges in deep water. 



Figures 8 to 11 will show the effects of shot depth on the 

 signals. The hydrophone depth and charge size were kept constant and 

 there is only a small variation of range. 



Figure 8 in this sequence shows oscillograms. Note that waves 

 corresponding to the first normal mode are weak for the 366-meter 

 shot. This is due to the pressure distribution with depth of the 

 first normal mode or, in terms of ray theory, the lack of waves 

 traveling along shallow-penetrating rays. 



Figures 9 and 10 show sound spectrograms for this series of 

 experiments. Up to seven normal modes are observed, but for the deep 

 shots the first normal mode is weak compared to the shallower shots. 

 The apparent cutoff for waves at 10 Hz is largely caused by a drop- 

 off of response of the listening system. 



Figure 11 shows the amplitude spectra of this series of shots 

 for the total signal (of duration 6 seconds) followed immediately 

 by 6 seconds of reverberation. Peaks at the bubble-pulse frequency 

 and harmonics are apparent on the narrow-band spectra. 



The effect of bathymetry on the signals is shown in Figure 12. 

 For this series of experiments, shots of identical size and depth 

 were launched from T-3 and recorded aboard ARLIS II employing a 

 hydrophone at a fixed depth. Note the weakening of the deep pene- 

 trating sounds as the amount of shallow-water along the path increases. 

 What this experiment shows is that the onset of a signal is largely 



696 



